U.S. patent number 5,120,908 [Application Number 07/607,537] was granted by the patent office on 1992-06-09 for electromagnetic position transducer.
This patent grant is currently assigned to Gazelle Graphic Systems Inc.. Invention is credited to Michael N. Gilano, Anthony E. Zank.
United States Patent |
5,120,908 |
Zank , et al. |
June 9, 1992 |
Electromagnetic position transducer
Abstract
Disclosed is electromagnetic transducer that does not require
complex analog to digital, digital to analog converters,
microprocessors or large memory circuitry. The transducer has a
helical transmitter coil with tap nodes spaced between its end
extremities. An oscillator drives the coil for simultaneously
producing oppositely directed current on opposite sides of selected
taps, creating a moving fringing field that is compatible with
modern graphic display tablet technology. A control circuit of the
transducer has a closed-loop integrator for detecting the centroid
of a received stylus signal. Analog and digital versions of the
integrator are disclosed.
Inventors: |
Zank; Anthony E. (Simi Valley,
CA), Gilano; Michael N. (Irvine, CA) |
Assignee: |
Gazelle Graphic Systems Inc.
(Irvine, CA)
|
Family
ID: |
24432711 |
Appl.
No.: |
07/607,537 |
Filed: |
November 1, 1990 |
Current U.S.
Class: |
178/19.04 |
Current CPC
Class: |
G06F
3/041 (20130101); G06F 3/047 (20130101); G06F
3/046 (20130101) |
Current International
Class: |
G06F
3/033 (20060101); G08C 021/00 () |
Field of
Search: |
;178/18,19 ;340/706 |
References Cited
[Referenced By]
U.S. Patent Documents
Other References
"Introduction to Graphic Digitizers" by W. Creitz and G. Helser;
GTCO Corporation, Columbia, Md.; Title page, pp. 3-43;
1986..
|
Primary Examiner: Schreyer; Stafford D.
Attorney, Agent or Firm: Sheldon & Mak
Claims
What is claimed is:
1. A position transducer comprising:
(a) helical electrically conductive transmitter coil having a
plurality of coil turns and first and second end extremities, and a
plurality of coil tap nodes thereon, the coil tap nodes being
spaced between the end extremities;
(b) oscillator means for producing a coil drive current;
(c) selector means for sequentially connecting the oscillator means
between the end extremities and selected ones of the coil tap nodes
for producing a moving fringing field resulting from portions of
the coil drive current flowing in opposite directions in the coil
on opposite sides of each of the selected nodes;
(d) a transducer body movable in a first transducer direction
relative to the transmitter coil and having a receiver coil fixedly
mounted thereto; and
(e) circuit means connected to the selector means and responsive to
the receiver coil for providing a first position signal, the first
position signal representing the position of the transducer body
relative to a first position reference, the first position
reference being perpendicular to the first direction.
2. The transducer of claim 1, wherein the coil drive current flows
simultaneously in opposite directions on opposite sides of the
selected nodes.
3. The transducer of claim 1, wherein the coil drive current flows
in opposite directions on opposite sides of the selected nodes in
alternate intervals.
4. The transducer of claim 1, wherein the transmitter coil
comprises oppositely helically wound coil components for returning
the coil drive current from the selected nodes to proximate an end
extremity of the transmitter coil, for suppressing stray magnetic
fields.
5. The transducer of claim 1, wherein the transmitter coil is
cylindrically helical about a coil axis and having front and back
portions on opposite sides of the coil axis, the coil axis being
parallel to the first transducer direction, the coil taps being
located on the back portion of the coil, the transducer further
comprising means for guiding the body at a fixed probe distance t
from the front portion of the transmitter coil.
6. The apparatus of claim 5, wherein the front and back portions of
the transmitter coil are spaced part by a winding distance d, the
distance d being at least approximately 0.03 inches.
7. The transducer of claim 5, wherein the turns of the transmitter
coil are substantially uniformly spaced.
8. The transducer of claim 5, wherein the receiver coil is
cylindrically symmetrical about a receiver axis, the transducer
body defining a stylus point on the receiver axis, the stylus point
moving at the probe distance t from the front portion of the
transmitter coil.
9. The transducer of claim 8, wherein the transmitter coil turns
are spaced in the direction of the coil axis by a turn spacing l,
the receiver coil having a diameter D and a length H, a central
point within the receiver coil on the receiver axis being offset by
a coil distance C from the stylus point, the coil distance C being
approximately C=a complex function of (D, l, H, d and t) for
permitting uniform operation of the transducer within a range of
inclinations of the receiver axis relative to orthogonal alignment
with the front portion of the coil.
10. The transducer of claim 9, wherein D.gtoreq.S.congruent.C.
11. The transducer of claim 9, further comprising means for gating
the circuit means,
whereby the circuit means is responsive to the receiver coil during
a sample interval only, a subset of the coil tap nodes being
activated by the selector means during the sample interval for
enhancing the uniformity of operation over the range of
inclinations of the receiver axis.
12. The transducer of claim 11, wherein the duration of the sample
interval is approximately equal to the time during which four of
the coil tap nodes are selected by the selector means.
13. The transducer of claim 8, wherein the receiver coil comprises
a pair of bi-filar wound receiver coil components, the circuit
means including a balanced differential input amplifier, the
receiver coil components driving the amplifier for producing a
receiver signal, the amplifier rejecting stray electrical
noise.
14. The transducer of claim 5, wherein the front coil portion is
substantially planar, the means for guiding the body comprising a
planar tablet surface for slidably supporting the body.
15. The transducer of claim 14, wherein the transmitter coil is a
first coil, the transducer comprising a second transmitter coil,
the second transmitter coil being orthogonally supported relative
to the first transmitter coil.
16. The transducer of claim 14, further comprising an electronic
graphic display unit, the display unit forming the tablet
surface.
17. The transducer of claim 16, wherein the display unit is
visually responsive to movement of the transducer body relative to
the tablet surface.
18. The transducer of claim 5, wherein the transmitter coil is a
first transmitter coil, the transducer comprising a second
transmitter coil, the second transmitter coil being orthogonally
supported relative to the first transmitter coil, the front
portions of the coils being substantially coplanar.
19. The transducer of claim 1, wherein the oscillator means
comprises an oscillator circuit having an oscillator output and a
reference voltage, the oscillator output driving the coil end
extremities, the reference voltage being sequentially connected to
at least some of the coil tap nodes by the selector means.
20. The transducer of claim 19, wherein the oscillator circuit
comprises a square wave oscillator.
21. The transducer of claim 20, wherein the end extremities of the
transmitter coil are connected to the oscillator circuit through a
constant current drive circuit, a low-pass frequency filter being
connected between the oscillator circuit and the constant current
drive circuit for shaping a smooth AC transmitter coil current
waveform.
22. The transducer of claim 20, comprising a pair of the
transmitter coils for dual-axis position measurements in
alternating axis intervals, the end extremities of the coils being
driven by common constant current sources through diode
isolators.
23. The transducer of claim 1, wherein the transmitter coil
includes a plurality of the turns between each of the coil tap
nodes.
24. The transducer of claim 1, wherein the circuit means comprises
a receiver node, a receiver signal at the receiver node being
responsive to the magnitude of the fringing field at the receiver
coil.
25. The transducer of claim 24, wherein the receiver node is
electrically connected to the receiver coil.
26. The transducer of claim 24, wherein the receiver node is
electrically connected to the transmitter coil, the receiver signal
being responsive to electric current loading of the transmitter
coil by the receiver coil.
27. The transducer of claim 24, comprising threshold means for
detecting a predetermined magnitude of the receiver signal, the
threshold means providing a valid signal when the transducer body
is aligned within a predetermined distance from the transmitter
coil.
28. The transducer of claim 27, wherein the oscillator circuit is
operable in a burst mode having a burst duty cycle, the burst duty
cycle being less than approximately 20 percent for conserving
electrical power.
29. The transducer of claim 28, wherein the burst mode is
terminated for at least a predetermined period of time upon
occurrence of the valid signal.
30. The transducer of claim 24, wherein the circuit means further
comprises:
(a) integrator means for summing a first-polarity component of the
receiver signal during a first cycle interval, the integrator means
also summing an opposite-polarity component of the receiver signal
during a second cycle interval;
(b) latch means for latching a variable position signal, the
position signal being representative of the location of the
sequentially connected coil tap nodes along the transmitter coil;
and
(c) feedback means for activating the latch means in response to
the integrator means, the first cycle interval terminating and the
second cycle interval commencing upon activation of the latch
means.
31. The transducer of claim 30, wherein the variable position
signal is generated by an N-state counter, N being a multiple M of
the number of the coil tap nodes of the transmitter coil, the
selector means decoding the N-state counter for connecting each of
the coil tap nodes to the oscillator means during an interval group
number of counter states, the interval group number corresponding
to the multiple M.
32. The transducer of claim 31, wherein the N-state counter is
clocked by the oscillator means.
33. The transducer of claim 31, wherein the multiple M is between 1
and 1024.
34. The transducer of claim 31, wherein the first and second cycle
intervals are limited to a total sample interval, the sample
interval corresponding to a number P of the counter states, the
number P being the multiple M multiplied by a sample factor, the
sample factor being from approximately 2 to approximately 5.
35. The transducer of claim 31, wherein the sample factor is
approximately 4.
36. The transducer of claim 30, wherein the integrator means
comprises a bidirectional counter and a variable frequency
oscillator, the frequency of the variable frequency oscillator
being responsive to the magnitude of the receiver signal.
37. A position transducer comprising:
(a) electrically conductive first and second transmitter coils,
each having a plurality of substantially uniformly spaced coil
turns and opposite end extremities, and a plurality of coil tap
nodes spaced between the end extremities, a plurality of the turns
being included between adjacent ones of the coil tap nodes, the
transmitter coils being cylindrically helical about respective
first and second coplanar coil axes and having front and back
portions on opposite sides of the coil axes, the coil tap nodes
being located on the back portions of the respective coils;
(b) oscillator means for producing a coil drive current
simultaneously in opposite directions between opposite sides of a
selected one of the coil tap nodes and the coil end extremities of
the corresponding coil;
(c) selector means for sequentially connecting the oscillator means
to different selected ones of the coil tap nodes for producing a
moving fringing field;
(d) a transducer body movable relative to the transmitter coils and
having a receiver coil cylindrically symmetrical about a receiver
axis and fixedly mounted to the transducer body, the transducer
body defining a stylus point on the receiver axis;
(e) means for guiding the body at a predetermined probe distance
from the front portions of the transducer coils;
(f) circuit means connected to the selector means and responsive to
the receiver coil for providing first and second position signals,
the position signals representing position coordinates of the
transducer body relative to the transmitter coils, the circuit
means comprising:
(i) a receiver node, a receiver signal at the receiver node being
responsive to the magnitude of the fringing field at the receiver
coil;
(ii) integrator means for summing a first-polarity component of the
receiver signal during a first cycle interval, the integrator means
also summing an opposite-polarity component of the receiver signal
during a second cycle interval;
(iii) latch means for latching a variable position signal
associated with each of the transmitter coils, each position signal
being representative of the location of the sequentially connected
coil tap nodes along the respective transmitter coil; and
(iv) feedback means for activating the latch means in response to
the integrator means, the first cycle interval terminating and the
second cycle interval commencing upon activation of the latch
means;
(g) means for gating the circuit means, whereby the circuit means
is responsive to the receiver coil during a sample interval only
for each transmitter coil, a subset of the coil tap nodes being
activated by the selector means during the sample interval for
enhancing linearity of the position signal between the coil tap
nodes; and
(h) threshold means for detecting a predetermined magnitude of the
receiver signal, the threshold means providing a valid signal when
the transducer body is aligned within a predetermined distance from
the first transmitter coil.
38. A method for measuring a coordinate position, comprising the
steps of:
(a) providing a helical transmitter coil having end extremities and
a plurality of coil tap nodes spaced between the end
extremities;
(b) driving the transmitter coil with an AC signal for producing
electrical current in the coil simultaneously in opposite
directions on both sides of a selected one of the coil tap
nodes;
(c) sequentially selecting a plurality of the coil tap nodes for
producing a moving fringing field;
(d) locating a receiver coil at the coordinate position for
producing a receiver signal responsive to the fringing field;
and
(e) detecting a centroid position of the receiver signal, the
centroid position being representative of the coordinate position.
Description
BACKGROUND
The present invention relates to position transducers such as
graphic tablets and the like for inputting absolute coordinate data
to complex control systems and other electronic equipment.
Position input devices such as joysticks, trackballs, and "mice"
are commonly used for feeding incremental data to a computer in
response to an operator's hand movements, the operator usually
adjusting the movements while observing a screen display of the
computer. In some important applications, it is desired to feed
coordinates that are directly measured from an existing exhibit
such as a drawing. In these applications the incremental input
devices have major shortcomings regarding both scale factor and
position reference.
Accordingly, a variety of graphic tablets have been developed for
generating computer position data from a workpiece such as a
drawing. Typical graphic tablets of the prior art provide absolute
position data relative to a fixed or adjustable reference, the data
being generated in response to a stylus that is moved by the
operator over the drawing surface. Among the important performance
parameters of such transducers are accuracy, repeatability,
resolution in both distance and time, range, limiting speed, and
ease of operation. Also important are cost, reliability, and
compatibility with related systems such as graphic displays.
An important class of graphic tablet incorporates an orthogonal
pair of conductor patterns within the tablet. The stylus carries a
coil from which a stylus signal is generated. See for example, U.S.
Pat. No. 3,975,592 to Carvey, which discloses an array of the
conductors in each pattern, the conductors being sequentially
energized for producing corresponding electric fields, one of the
fields being detected by the stylus for determining a coarse
position of the stylus, after which a subset of the conductors is
energized at a different rate. In successive cycles of the
sequential energization, a sampled counterpart of the stylus signal
is integrated positively, then negatively at double amplification.
A counter is latched when the integration output reaches zero, the
latched counter value being intended to represent the centroid of
the detected signal relative to the coarse position. However,
latched counter values fail to indicate the real-time centroid of
the stylus signal in that computations based on successive samples
are employed, the results being subject to error when there is
movement of the stylus between samples. Also, cumbersome and
expensive sample-and-hold circuitry is required.
Another problem with such systems is that the detected field
strength between the coarse positions is non-linear. U.S. Pat. No.
4,088,842 to Ikedo discloses nonlinear interpolation for detecting
intermediate positions of the stylus in a system having the stylus
excited and an array of planar pick-up coils in the tablet.
Nonlinear interpolators, however, are undesirably complex,
expensive to produce, and limited in accuracy. In fact, most
electromagnetic coordinate tablets of the prior art require
sophisticated signal processing by complex and expensive analog to
digital and digital to analog converters, microprocessors, and
computer memory. Further, most such tablets use strictly
2-dimensional conductor patterns that are expensive to produce,
even with printed circuit techniques. Also, in order to avoid
expensive interpolation errors, very fine coarse resolution is
resorted to, with consequent added circuit complexity and cost.
Another class of coordinate reading devices utilizes
magnetorestrictive material as a vibration transmission media. See,
for example, U.S. Pat. Nos. 4,216,352 to Chamuel and 4,273,954 to
Takeuchi et al. These and similar systems are subject to the
limitations of accurately measuring or converting from full scale
analog values.
If is often desirable to combine a graphic display capability in a
coordinate position transducer. Unfortunately, however, some
display devices can be adversely affected by strong electromagnetic
fields such as are produced by typical coordinate transducers.
Also, magnetorestrictive materials such as used in some coordinate
tablets are opaque, thus preventing the use of back-lighting,
etc.
Thus there is a need for a position transducer that provides high
accuracy, repeatability, range and resolution, that is easy to
operate, reliable and inexpensive to produce, yet is compatible
with low-cost graphic computer and display devices.
SUMMARY
The present invention is directed to an electromagnetic transducer
that meets this need without requiring complex analog to digital,
digital to analog converters, microprocessors or large memory
circuitry. The transducer includes a helical electrically
conductive transmitter coil having a plurality of coil turns and
first and second end extremities, and a plurality of coil tap nodes
spaced between the end extremities; oscillator means for producing
a coil drive current; selector means for sequentially connecting
the oscillator means between the end extremities and selected ones
of the coil tap nodes for producing a moving fringing field
resulting from portions of the coil drive current flowing in
opposite directions in the coil on opposite sides of each of the
selected nodes; a transducer body movable in a first transducer
direction relative to the transmitter coil and having a receiver
coil fixedly mounted thereto; and circuit means connected to the
selector means and responsive to the receiver coil for providing a
first position signal representing the position of the transducer
body relative to a first position reference that is perpendicular
to the first direction. The current flow between the selected tap
node and the opposite end extremities thereof can be simultaneous.
Preferably, the transmitter coil includes oppositely helically
wound coil components for returning the coil drive current from the
selected nodes to an end extremity of the transmitter coil, for
suppressing stray magnetic fields. Alternatively, the current flows
one direction during a first predetermined interval, then in the
opposite direction during a second predetermined interval. In
another alternative, the coil includes multiple, overlapping
oppositely wound helixes, the current flowing simultaneously in
opposite directions through coils wound in opposite directions,
thus enhancing field strength, and reducing or eliminating stray
current paths around the periphery of the tablet, resulting in
improved performance near the end extremities.
The transducer coil can be cylindrically helical about a coil axis
that is parallel to the first transducer direction, including front
and back portions on opposite sides of the coil axis, the coil taps
being located on the back portion of the coil, the transducer
further including means for guiding the body at a predetermined
probe distance from the front portion of the coil. Preferably the
front and back portions of the transducer coil are spaced part by a
winding distance d, the distance d being at least approximately
0.03 inches for avoiding undesirable magnetic flux concentration at
the coil end extremities. Preferably the turns of the transmitter
coil are substantially uniformly spaced for defining a constant
scale factor of the transducer.
The receiver coil can be cylindrically symmetrical about a receiver
axis, the transducer body defining a stylus point on the receiver
axis, the stylus point moving at the probe distance from the front
portion of the coil. Preferably, the coil turns are spaced in the
direction of the coil axis by a uniform turn spacing. Preferably a
central point within the receiver coil on the receiver axis is
offset by a coil distance from the stylus point, the coil distance
being approximately 0.4 inch for permitting uniform operation of
the transducer within a range of inclinations of the receiver axis
relative to orthogonal alignment with the front portion of the
coil. Preferably the transducer also includes means for gating the
circuit means, whereby the circuit means is responsive to the
receiver coil during a limited sample interval during which a
subset of the coil tap nodes are activated by the selector means
for enhancing the uniformity of operation over the range of
inclinations of the receiver axis. More preferably, the duration of
the sample interval is approximately equal to the time during which
four of the coil tap nodes are selected by the selector means. The
receiver coil can include a pair of bi-filar wound receiver coil
components, the circuit means including a balanced differential
input amplifier, the receiver coil components driving the amplifier
for producing a receiver signal, the amplifier rejecting stray
electrical noise.
The front coil portion can be substantially planar, the means for
guiding the body including a planar tablet surface for slidably
supporting the body. The transmitter coil can be a first coil, the
transducer including a second transmitter coil that is orthogonally
supported relative to the first transmitter coil. An electronic
graphic display unit can be included for forming the tablet
surface. The display unit can be visually responsive to movement of
the transducer body relative to the tablet surface. The front
portions of the transmitter coils can be substantially
coplanar.
The oscillator means can include an oscillator circuit having an
oscillator output and a reference voltage, the oscillator output
driving the coil end extremities, the reference voltage being
sequentially connected to at least some of the coil tap nodes by
the selector means. The oscillator circuit can include a square
wave oscillator. Preferably the end extremities of the transmitter
coil are connected to the oscillator circuit through a constant
current drive circuit, a low-pass frequency filter being connected
between the oscillator circuit and the constant current drive
circuit for shaping a smooth AC transmitter coil current waveform.
Dual-axis position measurements can be performed in alternating
axis intervals, the end is:extremities of the first and second
transmitter coils being driven by common constant current sources
through diode isolators.
Preferably the transmitter coil has a plurality of the turns
between each of the coil tap nodes for concentrating the fringing
field. The circuit means can include a receiver node, a receiver
signal at the receiver node being responsive to the magnitude of
the fringing field at the receiver coil. The receiver node can be
electrically connected to the receiver coil. Alternatively, the
receiver node can be electrically connected to the transmitter coil
for wireless operation of the receiver coil, the receiver signal
being responsive to electric current loading of the transmitter
coil by the receiver coil. Threshold means can be included for
detecting a predetermined magnitude of the receiver signal, the
threshold means providing a valid signal when the transducer body
is aligned within a predetermined distance from the transmitter
coil. The oscillator circuit can be operable in a burst mode having
a burst duty cycle that is less than approximately 20 percent for
conserving electrical power. The burst mode can be terminated for
at least a predetermined period of time upon occurrence of the
valid signal.
Preferably the circuit means further includes integrator means for
summing a first-polarity component of the receiver signal during a
first cycle interval and for summing an opposite-polarity component
of the receiver signal during a second cycle interval for locating
a centroid of the receiver signal, the circuit means also having
latch means for latching a variable position signal that is
representative of the location of the sequentially connected coil
tap nodes along the transmitter coil, and feedback means for
activating the latch means at the receiver signal centroid in
response to the integrator means, the first cycle interval
terminating and the second cycle interval commencing upon
activation of the latch means. The variable position signal can be
generated by an N-state counter, N being a multiple M of the number
of the coil tap nodes of the transmitter coil, the selector means
decoding the N-state counter for connecting each of the coil tap
nodes to the oscillator means during an interval group number of
counter states, the interval group number corresponding to the
multiple M. The N-state counter can be clocked by the oscillator
means. The multiple M can be nominally between 1 and 1024. The
first and second cycle intervals can be limited to a total sample
interval that corresponds to a number P of the counter states, the
number P being the multiple M multiplied by a sample factor, the
sample factor being from approximately 2 to approximately 5.
Preferably the sample factor is approximately 4 for enhancing the
uniformity and linearity of the transducer.
The integrator means can include a bidirectional counter and a
variable frequency oscillator, the frequency of the variable
frequency oscillator being responsive to the magnitude of the
receiver signal.
In another aspect of the present invention, a method is disclosed
for measuring a coordinate position, including the steps of:
(a) providing a helical transmitter coil having end extremities and
a plurality of coil tap nodes spaced between the end
extremities;
(b) driving the transmitter coil with an AC signal for producing
electrical current in the coil simultaneously in opposite
directions on both sides of a selected one of the coil tap
nodes;
(c) sequentially selecting a plurality of the coil tap nodes for
producing a moving fringing field;
(d) locating a receiver coil at the coordinate position for
producing a receiver signal responsive to the fringing field;
and
(e) detecting a centroid position of the receiver signal, the
centroid position being representative of the coordinate
position.
DRAWINGS
These and other features, aspects, and advantages of the present
invention will become better understood with reference to the
following description, appended claims, and accompanying drawings,
where:
FIG. 1 is a combination perspective and simplified schematic
diagram of position transducer apparatus according to the present
invention;
FIG. 2 is a sectional view showing a magnetic field configuration
of the apparatus of FIG. 1;
FIG. 3 is a schematic diagram showing an alternative configuration
of a portion of the apparatus of FIG. 1;
FIG. 4 is a schematic diagram showing an alternative configuration
of the apparatus of FIG. 1;
FIG. 5 is a schematic diagram showing an alternative configuration
of a circuit portion of the apparatus of FIG. 4;
FIG. 6 is a schematic diagram of an alternative configuration of a
portion of the circuit of FIG. 5;
FIG. 7 is a timing diagram for the apparatus of FIG. 4;
FIG. 8 is a schematic diagram of an alternative configuration of a
portion of the circuit of FIG. 1;
FIG. 9 is a combination perspective and simplified schematic
diagram of a portion of the apparatus of FIG. 1;
FIG. 10 is a simplified schematic diagram showing an alternative
configuration of a circuit portion of the apparatus of FIG. 4;
and
FIG. 11 is a schematic diagram shown an alternative configuration
of a portion of the circuit of FIG. 4.
DESCRIPTION
The present invention is directed to an electromagnetic transducer,
an exemplary configuration of which forms a graphic tablet that
provides digital coordinate data for a computer or similar system
in response to operator control. With reference to FIGS. 1 and 2 of
the drawings, a transducer apparatus 10 according to the present
invention includes a 3-dimensional coil assembly 12, a tablet
surface 14 being supported in fixed parallel relation to a major
coil plane 16 of the coil assembly 12 for movably supporting a
stylus assembly 18 at a predetermined tablet distance t from the
major coil plane 16. The tablet surface 14, as further described
below, can receive a drawing 20 or the like, the stylus assembly 18
being manually movable upon features of the drawing 20 for
measurement thereof.
The coil assembly 12 includes a pair of helical coils 22,
designated 22.sub.X and 22.sub.Y, each of the coils 22 having a
plurality of taps 24 and a pair of end extremities 26. As used
herein, the term "helical" is used in its broad sense, meaning
generated by a point that moves with a distance component parallel
to a line, the distance component increasing in one direction only
as the point orbits the line, the point advancing a minimum
distance in the parallel direction for each orbit of the line. In a
preferred configuration of the coils 22 as depicted in the
drawings, the coils 22 are also cylindrical in that the turns of
each coil lie on a surface that is generated by a line that moves
parallel to a stationary line or coil axis 27. The turns of each
coil 22 each have a top portion 28 substantially in the coil plane
16 and a bottom portion 30 uniformly spaced below the coil plane
16, the coil axis 27 being located midway between the top and
bottom portions 28 and 30, the portions 28 and 30 being spaced
apart by a coil depth d. The top portions 28 are parallel and
uniformly spaced apart by a turn spacing l, each of the bottom
portions 30 being located midway between neighboring top portions
28 and parallel thereto. Each coil 22 also includes a plurality of
obliquely oriented side portions 32 that serially connect the top
and bottom portions 28 and 30. The top portions 28 and the bottom
portions 30 of the respective coils 22.sub.X and 22.sub.Y are
orthogonal for permitting independent measurement of corresponding
X and Y position coordinates of the stylus assembly 18, X and Y
coordinate directions being indicated by the arrows in FIG. 1.
The taps 24 are preferably provided at the edges of at least some
of the coil bottom portions 30. In the configuration of FIGS. 1 and
2, every third bottom portion 30 of each coil 22 has one of the
taps 24, the taps 24 having a tap spacing S=3l. As best shown in
FIG. 2, adjacent turns of the coils 22 on opposite sides of the
taps 24 produce an electromagnetic fringing field 34, having right
and left flux paths 34R and 34L, when energized as described
herein, the field 34 being concentrated between adjacent ones of
the top portions 28 on opposite sides of a selected one of the taps
24. A stylus coil 36 is provided within the stylus assembly 18, the
coil 36 being concentric with a stylus axis 37 of the assembly 18.
The stylus coil 36 is responsive to the fringing field 34 for
producing a stylus signal 38, also further described herein, when a
stylus point 39 that is formed on the axis 37 at a lower extremity
of the stylus assembly 18 is in proximate contact with the tablet
surface 14.
With particular reference to FIG. 1, the coil assembly 12 and the
stylus assembly 18 are connected to a feedback control circuit 40
of the apparatus 10. The control circuit 40 includes an oscillator
42 for AC drive of each of the coils 22, a buffered output 44 from
the oscillator 42 being connected to each of the end extremities 26
of the coils 22 through a suitable passive coil terminator 46, a
pair of the terminators 46 being included in an X axis portion
48.sub.X of the control circuit 40 for driving the coil 22.sub.X.
The control circuit 40 also includes a Y axis counterpart 48.sub.Y
(not shown) of the circuit portion 48.sub.X. The taps 24 of each
coil 22 are sequentially driven by a demultiplexer or decoder 50
that is included within the axis portion 48.sub.X for driving a
selected one of the taps 24 to a reference or ground potential,
outputs of the decoder 50 having appropriate power handling
capacity for passing a desired level of current through the coil
assembly 12. The fringing field 34 is thus produced by electrical
current that flows to the selected tap 24 in opposite directions
from the end extremities 26 of the associated coil 22. The purpose
of the terminators 46 is to reduce high frequency electromagnetic
radiation from the coils 22 by filtering harmonics from the current
of the buffered output 44 from the oscillator 42. Also, the output
44 can be a conveniently obtained square wave voltage, with a
substantially sine-wave current profile being produced in the coil
22. The terminators 46 also serve to balance the current in the
coil 22, equalizing the current between the selected tap 24 and the
opposite end extremities 26.
With further reference to FIG. 3, an alternative configuration of
the control circuit 40 has the oscillator 42 being selectively
connected to the taps 24 through the decoder 50, the coil
terminators 46 being connected to a reference such as ground as
depicted in FIG. 3. The fringing field 34 in this alternative
configuration has the same form as described above in connection
with FIG. 2 because the current distribution in the coil 22 is the
same except for flowing in opposite directions from the case of the
circuit configuration of FIG. 1. Advantageously, the configuration
of FIG. 1 permits the decoder 50 to be arranged more simply than
for the alternative configuration, being required only to sink
current at ground potential from the selected tap 24.
The oscillator 42 also provides timing for the control circuit 40,
the oscillator output 42 clocking a binary counter 52 having a
plurality of counter outputs 54 in two groups. The most significant
counter outputs, designated 54.sub.Hi, drive corresponding inputs
of the decoder 50 for sequentially selecting the taps 24, whereby
the fringing field 34 is caused to move stepwise along each of the
coils 22. The counter outputs 54.sub.Hi, together with the other
counter outputs, designated 54.sub.Lo, are also fed to a latch
circuit 56 for storing X coordinate data that is produced by the
apparatus 10 as described herein.
As indicated above, the stylus signal 38 is responsive to the
fringing field 34 according to the position of the stylus assembly
18 relative to the coil assembly 12. The stylus signal 38 is fed to
a stylus amplifier 58, from which an AC sensor output 60 is
connected to a demodulator 62 and low-pass filter 64 for producing
a DC sensor output 66 for driving a closed-loop centroid null
circuit 68 according to the present invention. The null circuit 68
includes a phase splitter 70, a polarity selector 72, an integrator
amplifier 74, a one-shot or timer 76 having a latch output 78, and
the latch circuit 56, the coordinate data present at the latch
circuit 56 being stored therein upon activation of the latch output
78. The integrator amplifier 64 and the timer 76 are also included
with the latch circuit 56 in the axis portion 48.sub.X of the
control circuit 40. The polarity selector 72 feeds the integrator
amplifier 74 alternately with in-phase and inverted counterparts of
the DC sensor output 66, also in response to the latch output 78 of
the timer 76, the timing duration of the timer 76 being variable in
response to the output of the integrator amplifier 74. Operation of
the polarity selector 72 by the timer 76 is such that the signal to
the integrator amplifier 74 is balanced about a null reference
level, the latch output 78 being activated in successive
integration cycles of the null circuit 68 for latching the
coordinate data and reversing the direction of integration at the
centroid of the DC stylus signal 66.
Once latched, and until the next integration cycle, the latch 56
holds a binary representation of the position of the stylus
assembly 18 relative to the end extremity 26 adjacent to the
first-selected tap 40 of the coil 22. In continuing operation of
the control circuit 40, the axis portions 48.sub.X and 48.sub.Y
alternately store the X and Y coordinate positions of the stylus
assembly 18.
An important advantage of the coil assembly 12 of the present
invention is that the centroid location of the stylus signal 38 is
a highly linear function of the coordinate position of the stylus
point 39 on the stylus surface 14 relative to the position of the
selected tap 24. As the stylus point 39 approaches the selected tap
24, the amplitude of the stylus signal 38 increases linearly until
the point 39 passes above the tap 24, then the signal 38
symmetrically decreases as the point continues away from the tap
24. Thus there is no need for complex non-linear interpolation
hardware or software in the present invention. Whereas the
uncompensated linearity between coarse positions of typical prior
art tablet transducers having planar conductor patterns is only
about 50%, the coil assembly 12 of the present invention provides a
linearity of approximately 1% or better between adjacent taps
24.
The coil assembly 12 can advantageously incorporate a graphic
display unit 790 that forms the tablet surface 14, the display unit
790 having a flat display panel 792 that provides, for example, a
visual display of the path of the stylus point 39 on the tablet
surface 14. The fringing field 34 penetrates the display panel 792
without interfering with electronics therein, and the apparatus 10
can include or operate in conjunction with a metalized or
electro-luminescent backlight assembly (not shown). Advantageously,
special shielding is not required between the display panel 792 and
other components of the apparatus 10, because the field 34 is of
the fringing type. Most of the field 34 is contained within the
coil assembly 12, and the lack of special shielding allows the
control circuit 40 and other electronics to be located immediately
behind or in the same plane as the coil assembly 12 along one edge
thereof.
Several methods exist for minimizing stray electromagnetic fields
around the tablet periphery or underside. These include shielding
of current return wires, placing return wires on the back side of
the tablet, returning the current through "return" helixes
overlayed on the primary helix, and dividing the return current
equally between driven return helixes. This helix return method
totally eliminates stray current flow. Other current return options
include multiplexing current flow in the tablet to be
non-simultaneous, controlled by fixed timing or by variable timing
tied to the position of the stylus pick-up. These methods, used
singly or in combination, are effective for enlarging the useful
area of the transducer 10, and for avoiding undesirable
interference with other nearby electronic equipment. One
particularly advantageous alternative configuration of the coil 22
is described below in connection with FIG. 9.
With further reference to FIG. 4, an alternative configuration of
the control circuit 40 has the null circuit 68 of FIG. 1
implemented digitally. In an exemplary and preferred digital
configuration, the timer 76 is replaced by a down-counter 80 that
is clocked directly by the oscillator 42 and periodically loaded
with position data; the integrator amplifier 74 is replaced by
respective up/down counters 82, designated 82.sub.X and 82.sub.Y in
FIG. 4, the counters 82 generating the position data for the
down-counter 80; and the polarity selector 72 is replaced by
direction controls CDN for the up/down counters 82. A clock divider
84 replaces the binary counter 52, the divider 84 being operated at
2.0 MHz by a system clock signal SC from the oscillator 42, the
divider 84 having a demodulator output DM at 500 KHz for operating
a synchronous demodulator (described below) and for driving the
coils 22 as described below. The divider 84 also generates a shift
register clock SRC at a submultiple of the DM output frequency, and
a carry or timer load output /TLD at a submultiple frequency of the
clock SC. Also, the decoder 50 is replaced by a coil drive shift
register 86, the shift register 86 being clocked at 15,625 Hz by
the clock SC of the clock divider 84. A single bit of the shift
register 86 is periodically activated at approximately 2 ms
intervals by the output /TLD of the clock divider 84, the other
bits of the shift register 86 being sequentially activated in
successive cycles of the clock SC. In the exemplary configuration
of FIG. 4, there are 32 outputs of the shift register 86 for
sequentially activating up to 32 of the taps 24 of each of the
coils 22.
A coil buffer circuit 88 receives the DM output (square wave) from
the clock divider 84, the buffer circuit 88 including a low-pass
filter 90 and a pair of current waveform drivers 92 for producing
an approximately sine-wave current into the opposite end
extremities 26 of the tablet coils 22. A pair of steering diodes 94
is provided at each of the current drivers 92 for isolating the end
extremities 26 of the tablet coils 22.sub.X and 22.sub.Y.
A tap driver circuit 96 is connected between the taps 24 of the
tablet coils 22 and the outputs of the shift register 86 for
sinking to ground the coil current from the selected tap 24. The
tap driver circuit 96 includes a plurality of tap driver
transistors 98, each driver transistor 98 being connected between
one of the taps 24 and an appropriate ground select line 100, there
being a pair of the select lines 100.sub.X and 100.sub.Y associated
with the respective tablet coils 22.sub.X and 22.sub.Y. The shift
register 86 has at least as many stages as the number of taps 24 on
the longest of the coils 22. Each of the ground lines 100 is
selectively connected to ground by a grounding transistor 102 in
response to corresponding axis enable signals X and Y (described
herein), the grounding transistors being designated 102.sub.X and
102.sub.Y. In configurations of the apparatus 10 wherein the number
of stages of the shift register 86 matches the number of taps 24 on
each of the coils 22.sub.X and 22.sub.Y, there would be a pair of
the driver transistors 98 activated from each output of the shift
register 86, one of the transistors 98 sinking current from a tap
24 of one tablet coil 22 and the grounded select line 100. The
tablet coils 22.sub.X and 22.sub.Y are driven in alternate cycles
of the clock divider 84, the output /TLD driving an axis divider
104 for producing complementary cycle outputs X and Y, the cycle
outputs X and Y activating, respectively, the grounding transistors
102.sub.X and 102.sub.Y.
The up/down counters 82.sub.X and 82.sub.Y are driven at variable
frequency by a variable frequency oscillator (VCO) 106 in response
to the sensor outputs 60 and 66 as described herein. As indicated
in FIG. 4, the stylus amplifier 58 is implemented as an AC coupled
pair of high-speed operational amplifiers 108, designated 108.sub.A
and 108.sub.B, the stylus coil 36 being connected between a
passively generated reference voltage VR and a non-inverting input
of the first amplifier 108.sub.A, the voltage VR also being used as
a reference for the second amplifier 108.sub.B, the AC sensor
output 60 being produced by the second amplifier 108.sub.B. A
synchronous demodulator 110 is used in place of the demodulator 62
of FIG. 1 for rejecting out of phase components of the AC sensor
output 60. The demodulator 110 feeds a counterpart of the low-pass
filter 64, designated 112 in FIG. 4, the filter 112 being
relatively fast for rapid response of the DC sensor output 66 to
the AC sensor output 60. The DC sensor output 66 is fed to a
comparator circuit 114 for charging a capacitor 116 at a rate
proportional to the extent the DC sensor output 66 (active low)
goes a predetermined amount below a reference voltage 118, the
reference voltage 118 being produced by a passive low frequency
filter 120 that is driven by the AC sensor output 60. Thus the
reference voltage 118 represents an average DC value of the AC
sensor output 60, the filter 120 having a much slower response than
the low-pass filter 112. The comparator circuit 114 is very
sensitive to changes in the DC sensor output 66, but compensates
for DC offset drift such as might be introduced by the amplifier
108.sub.B because the low frequency filter 120 is resistively
coupled to the AC sensor output 60. An exemplary and preferred
configuration of the low frequency filter 120 has a response time
on the order of 1 ms, while the low-pass filter 112 has a response
time of 0.025 ms, successive ones of the taps 24 being activated at
intervals of approximately 2 ms.
The VCO 106 is operative over a frequency range of from DC to above
1 MHz, having an output inverter 122 that is driven by a fast
(Baker-clamped) comparator 124, the comparator 124 being responsive
to a voltage V at the capacitor 116 and having positive feedback
from a feedback inverter 126 that is fed by the VCO output from the
inverter 122. The output of the feedback inverter 126 is also
connected to the capacitor 116 through a diode 128, the diode 128
providing a discharge path for the capacitor 116, the voltage V at
the capacitor 116 increasing when a current in excess of a
threshold current is produced by the comparator circuit 114, and
decreasing when the VCO output of the inverter 122 is high. Also,
the VCO 106 is gated by a VEN signal (described below) that is
connected to the capacitor 116 by a diode 130, the VCO output of
the inverter 122 being held at ground when the VEN signal is low,
the diode 128 then preventing the voltage V at the capacitor 116
from rising. Thus the VCO 106 clocks the counters 82 at rates
proportional to the magnitude of the DC sensor output 66, whenever
the output 66 is sufficiently large and the VEN signal is also
high.
The VEN signal is arbitrarily held high until a lock signal Z
(further described below) is activated for permitting the counters
82 to stabilize upon initial entry of the stylus assembly 18 within
the fields 34 of the coil assembly 12. When the lock signal Z is
active, the VEN signal is responsive to an enable flip-flop 132
that is momentarily set during a terminal count sequence of the
down counter 80 by a /TOUT signal from a "carry" output of the
counter 80, the flip-flop 132 being clocked by an ST output from an
intermediate stage of the counter 80. In the exemplary circuit
configuration of FIG. 4, the ST output is from the ninth stage of
the (12 stage) counter 80, thus activating the VEN signal for 256
.mu.s following activation of the /TOUT signal, 256 .mu.s
corresponding to the period for four tap selections by the shift
register 86. A /PO signal is activated during the second half cycle
of the ST output that activates the enable flip-flop 132, the /PO
signal resetting a first direction flip-flop -34, a DN signal from
the flip-flop 134 being fed to a second direction flip-flop 136 for
producing the direction signal CDN synchronously with the VCO
output. Thus the counters 82 count downwardly during the first half
of VEN (when the lock signal Z is active), then upwardly during the
second half of VEN. Thus once lock is achieved, the VCO 106
operates only during an interval of time that is equivalent to the
duration of four selections of taps 24 by the shift register 86 and
the associated tap driver circuit 96. Accordingly, in each axis
cycle (X or Y) beginning with activation of the /TLD output of the
counter 84, the counter 82 retains its prior value until the /TOUT
signal is activated by underflow of the down-counter 80, the
down-counter 80 having previously been set with the contents of the
counter 82 by the /TLD signal for delaying the VEN signal until
approximately two tap selections prior to a time associated with a
previously determined position of the stylus assembly 18. Selection
between the data from the counters 82.sub.X and 82.sub.Y to be
loaded into the down-counter 80 is effected by an axis selector 138
in response to the X signal from the axis divider 104, whereby the
down-counter 80 is loaded with X and Y position data in alternate
axis cycles. Upon activation of the VEN signal, and in the presence
of a sufficiently large output from the stylus 18 as described
above, the counter 82 commences counting down at a rate
proportional to the DC sensor output 66 until the CDN signal goes
low (on the first VCO pulse after the midpoint of the VEN signal)
at which point the counter 82 counts upwardly until termination of
the VEN signal, at which point the counter 82 holds a new
coordinate position measurement for that axis.
Counterparts of the latch circuit 56, designated 56.sub.X and
56.sub.Y in FIG. 4, are loaded with the X and Y coordinate data
from the counters 82.sub.X and 82.sub.Y subsequent to termination
of the respective count sequences. Conveniently, this can be
effected for each axis during activation of the other axis.
Arbitrarily, the midpoint of the inactive intervals is used for
loading the latch circuits 56, respective XLD and YLD strobe
signals for the latch circuits 56 being enabled by an SR16 output
of the shift register 86, gated with the opposite Y and X signals
from the axis divider 104. In the circuit of FIG. 4, each of the
latch circuits 56 is implemented with 16 data bits, including 12
bits of position data, the lock signal Z, and a pen down signal
PDN, two unused bits being available for future use. The signal PDN
is generated by a switch circuit 140 that is connected to a
push-button switch 142 on the stylus assembly 18, the switch 142
providing a mouse "click" or similar function whereby an operator
of the apparatus 10 can signal the location of the stylus point 39
at particular features of the drawing 20, for example.
Operation of the control circuit 40 of FIG. 4 is illustrated in the
exemplary timing diagram of FIG. 7, wherein the AC sensor output 60
is labeled "AC". As shown in FIG. 7, AC is active during six tap
selections of the coil 22.sub.X (X active) immediately following
activation of SR16, and during six tap selections of the coil
22.sub.Y (X inactive). In this example, the stylus point 39 would
be slightly to the right of midway between the 20th and 21st taps
24 in the X direction of the coil 22.sub.X. Accordingly, the AC
signal bursts are shown as being asymmetrical about the tap
selection transitions of the shift register clock SRC, the burst
during X being concentrated to the right of the transition to the
20th taps 24 of the coil 22.sub.X, the other burst being
concentrated to the left of the transition to the 8th tap 24 of the
coil 22.sub.Y slightly below midway between the 7th and 8th taps 24
in the Y direction of the coil 22.sub.Y.
In the above example, and assuming that the lock signal Z is
active, the initiation of the /PO signal (and the midpoint of the
VEN signal) follows the centroid of that portion of the stylus
signal 138 that lies within the period of activation of the VEN
signal. Thus the stylus signal 38 is advantageously ignored when
the selected tap 24 is more than two positions from the stylus
point 39, thereby facilitating uniform response of the apparatus 10
to positions of the stylus point 39 approaching the end extremities
26 of the coils 22. As also shown in FIG. 4, the coil buffer
circuit 88 is gated with the VEN signal for significant power
savings while the lock condition is maintained. For example, in the
32-bit configuration of the drive shift register 36, the power
consumption by the coil 22 is reduced to approximately one-eighth
when lock is achieved. Even using the 8 tap configuration of the
coil 22 shown in FIG. 1 with an 8-bit configuration of the shift
register 86, an approximately 50 percent power savings is
possible.
More importantly, the gating of the stylus signal 38 as described
above advantageously enhances the tolerance of the apparatus 10 to
variations of the stylus axis 37 from being normal to the tablet
surface 14. As shown in FIG. 2, the stylus coil 36 has a diameter D
(which is preferably greater than the turn spacing l) and a coil
height H, the midpoint of which is located a coil distance C from
the stylus point 39. An experimental prototype of the apparatus 10
has been built and tested, the relevant dimensions of the stylus
assembly 18 being approximately D=0.25 inch, H=0.2 inch (200 turns
of wire), and C=0.40 inch. Similarly, the coil assembly 12 of the
experimental prototype was fabricated with the relevant dimensions
being approximately l=0.10 inch, S=0.30 inch, d=0.10 inch, and
t=0.20 inch. Although the stylus signal 38 is somewhat insensitive
to variations in the orientation of the stylus axis 37 from being
perpendicular to the tablet surface 14, when the VCO 106 is gated
as described above, there is a marked improvement in the immunity
of the control circuit 40 to such variations.
Based on further testing of and modifications to the experimental
prototype, it is believed that when the VCO 106 is enabled for a
period corresponding to four cycles of the shift register clock
SRC, a high degree of immunity to tipping of the stylus assembly 18
is achieved approximately when D.gtoreq.S.congruent.C.
As discussed above, the VEN signal to the VCO 106 is forced high
continuously until occurrence of the lock signal Z. As further
shown in FIG. 4, a pair of lock flip-flops, designated 144.sub.X
and 144.sub.Y, are set during corresponding axis intervals when at
least a predetermined number of pulses are produced by the VCO
prior to underflow of the down-counter 80. This condition is
produced when the DC sensor signal 66 is significantly large for
sufficiently charging the capacitor 116 above the current threshold
discussed above in connection with the comparator 114 when a coil
tap 24 is selected that is approximately two tap positions from the
stylus point 39. The occurrence of the predetermined number of
pulses is detected by a 8-bit lock shift register 146 that is
clocked by every fourth VCO pulse prior to occurrence of the/TOUT
signal by a VCO divider 148; thus the predetermined number of VCO
pulses is 32 for occurrence of the lock signal Z.
With further reference to FIGS. 5 and 6, a wireless configuration
of the stylus assembly 18 has the stylus coil 36 shorted through a
shunt reactance, which can be a shunt capacitor 150 as shown in
FIG. 5, or a direct shorting connection. In this configuration,
current induced in the stylus coil 36 loads the transmitter coil
22. The stylus signal 38 is generated by a current sensing resistor
152 that is connected between each grounding transistor 102 and
ground, variations in the current from the selected taps 24
resulting from the proximity of the stylus coil 36 to the selected
tap 24. As further shown in FIG. 5, a counterpart of the push
button switch 142 is connected across a portion of the stylus coil
36 for permitting an appropriate counterpart of the switch circuit
140 (not shown) to generate the PDN signal in response to operation
of the push button switch 142 by phase discrimination.
Alternatively a variable resistance counterpart of the push button
switch 142, designated pressure sensor 154, can be connected in
series with the shunt capacitor 150 as shown in FIG. 6, the
pressure sensor 154 being responsive to the pressure between the
stylus point 39 and the tablet surface 14. The mechanical
connection between the pressure sensor 154 and the stylus point 39
can be by any means known to those skilled in the art.
Alternatively, and with further reference to FIG. 8, improved
rejection of stray noise and electrical fields is achieved using a
dual counterpart of the stylus coil 36, coil portions 36a and 36b
being wound bi-filar and connected through a shielded cable 155 to
the stylus amplifier 58, the amplifier 58 being configured as a
balanced differential input amplifier for producing the AC sensor
output 60 as shown in FIG. 8.
With further reference to FIG. 9, an alternative configuration of
the coil assembly 12 has the transmitter coils 22 formed with
oppositely wound winding components 22a and 22b, only one
transducer axis being depicted for clarity. The winding component
22a is similar in form to the winding 22.sub.X of FIG. 1 and having
end terminations 26a and tap nodes 24a, the component 22b (shown by
dashed lines) being wound in an opposite direction and having end
terminations 26b and tap nodes 24b. Tap selection in the coil
configuration of FIG. 9 may be effected in a variety of ways, FIG.
9 further illustrating one preferred selection circuit. In
particular, the end terminations 26a are driven from the coil
buffer circuit of FIG. 4, the end terminations 26b are grounded
through counterparts of the coil terminators 46 of FIG. 1, and
selected tap nodes 24a and 24b are shorted together by counterparts
of the grounding transistors 102, designated selector transistors
102ab.
In the coil configuration of FIG. 9, current flows simultaneously
from the buffer circuit 88 through the end extremities 26a in
opposite directions toward the selected tap node 24a, then in
opposite directions away from the selected tap node 24b, through
the end extremities 26b, thence through the coil terminators 46 to
ground. The coil configuration of FIG. 9 advantageously generates a
desired magnitude of the fringing field 34 using only approximately
half of the coil current required for the coil configuration of
FIG. 1. Also, wiring connections to the end extremities 26 and the
tap nodes 24 carry current in opposite directions along parallel
paths for cancellation of stray fields that would otherwise be
produced by the wiring. Moreover, the oppositely wound coil
components 22a and 22b themselves produce the fringing field 34
more uniformly across the tablet surface 14 of the coil assembly
12.
With further reference to FIG. 10, in configurations of the control
circuit 40 wherein the coil 22 is driven by a constant current, as
in the configurations of FIGS. 4 and 9, wireless operation of the
stylus assembly 18 can be effected by sensing variations in the
voltage across the coil 22, instead of sensing variations in the
current as in the configuration of FIG. 5. As shown in FIG. 10, a
voltage divider 156 is connected between the end extremities 26
(26a in FIG. 9) of the coil 12 for sensing an average coil voltage,
the stylus amplifier 58 comparing the average coil voltage with a
counterpart of the oscillator output, designated OSC, for sensing
variations in loading of the transmitter coil 22 by the stylus coil
36. As shown in FIG. 10, a decoupler 157 is provided between the
filter 90 and the current waveform driver 92 of the coil buffer
circuit 88 for producing the OSC signal.
With further reference to FIG. 11, instead of the transistor coil
current being simultaneously driven in opposite directions toward
or away from the selected tap 24, the current can be driven one
direction during a first predetermined interval, then in the
opposite direction during a second predetermined interval. For this
purpose, the current waveform drivers 92 of the coil buffer circuit
88 are gated with corresponding interval transistors 92a, one of
the transistors 92a being driven in response to the DN signal, the
other transistor 92a being driven in a complementary interval by a
/DN signal from the first direction flip-flop 134. In FIG. 11 the
DN and /DN signals are also gated with the lock signal Z for
enabling alternate interval activation of the current waveform
drivers 92 once the lock condition is achieved. Thus the current in
each transmitter coil 22 flows from one end extremity 26 to the
selected tap node 24 during the first half of the VEN interval,
then the current flows from the other end extremity 26 in the
opposite direction toward the selected tap node 24 during the
second half of the VEN interval, because the steering diodes 94 are
cross-connected to the coil extremities 26 of the transmitter coils
22.sub.X and 22.sub.Y as shown in FIGS. 4 and 11.
In the version of the control circuit 40 shown in FIG. 1, the
decoder 50 can be implemented with generic 54/74 series integrated
circuit logic such as a CMOS 74C42 decoder, the outputs of which
are connected through readily available non-inverting tri-state
drivers to corresponding ones of the coil taps 24. Similarly, the
counter 52 can be implemented as three or more cascaded 74C161
binary counters, and the latch 56 can be implemented with any of
many circuits known to those having skill in the art that are
readily commercially available from a variety of sources.
In the version of the control circuit 40 shown in FIG. 4, the
down-counter 80, the up-down counters 82, the counter 84, and the
counter 148 can each be implemented with appropriate numbers of
generic 4 bit 54/74 series '191 up/down counter integrated
circuits. Similarly, the shift registers 86 and 146 can be
implemented with '164 8-bit shift register circuits. The selector
138 can be implemented with a '157 quad 2-input multiplexer, and
the axis divider 104 and the flip-flops 132, 134, 136, and 144 can
be implemented with '74 dual D flip-flop circuits. The bipolar
transistors, including bipolar transistors within the operational
amplifiers 104 can be 2N2222 (NPN) and 2N2907 (PNP) types.
The tablet coils 22 can be formed on a plastic core 158, being
wound with fine copper wire such as 36 AWG enameled coil wire.
Proper alignment of the coil portions can be facilitated by
locating grooves appropriately on the core 158. Alternatively, the
tablet coils 22 can be formed using methods known to those skilled
in the art of multi-layer circuit boards. These include printed
conductors, using 1, 2, 3, or 4 layers and etched conductors using
1, 2, 3, or 4 layers. Implementation of 2 layer dual-coil circuits
can be achieved also by slight coil distortion required to
interleave traces from one side of the printed circuit assembly to
the other side and back again. Any electromagnetic distortion
caused by interleaving coils can be easily corrected by correction
and table look-up methods commonly known to those skilled in the
art.
Although the present invention has been described in considerable
detail with reference to certain preferred versions thereof, other
versions are possible. For example, there are several ways to
implement analog and digital up/down integrators using operational
amplifiers, transistors, timers, processors, analog switches, and
combinations of such electronics. Similarly, there are several
alternative ways to drive the tablet coil, including passive,
active, constant current, AC drive at the ends, DC drive at the
ends and AC drive modulation at the return, etc. The taps 24 can be
connected to the return path via transistors, shift registers, or
other electronic means. Therefore, the spirit and scope of the
appended claims should not necessarily be limited to the
description of the preferred versions contained herein.
* * * * *